Dimensional crossover of in-plane thermal transport in few-layer graphene: intrinsic layer-number dependence, structural condition, and quantitative theory

The room-temperature in-plane thermal conductivity $K$ of suspended few-layer graphene was measured by Ghosh et al. (2010) to fall from roughly 2800-3000 to 1300-1500 W/mK as the number of atomic layers increases from 2 to 4, with the bulk graphite basal-plane value recovered near $n\approx 8$; the reduction was attributed to cross-plane coupling of low-energy phonons that enlarges the three-phonon Umklapp scattering phase space. The question this raised is still open. The intrinsic $K(n)$ itself is contested: an independent thermal-bridge measurement found only about 300-600 W/mK with no clear layer dependence for 2-4 layers; the structural condition (layer number together with stacking order and interlayer registry) that marks the intrinsic two-dimensional to three-dimensional crossover is unsettled, with molecular-dynamics studies finding registry effects unimportant while experiments find strong stacking and twist-angle dependence; and the available theory stops at three-phonon scattering even though four-phonon processes are known to reduce graphene's intrinsic conductivity substantially. The problem is to determine the intrinsic $K(n)$ as a function of layer number, identify the structural condition and phonon-scattering mechanisms of the crossover, and quantitatively separate extrinsic sample and measurement effects, so that a microscopic theory reproduces the measured crossover.

Background

Heat conduction in an electrically modest conductor such as graphite is carried almost entirely by phonons, the quantized vibrations of the crystal lattice, and is measured by the thermal conductivity $K$ appearing in Fourier's law, which relates the heat flux to the temperature gradient. Near room temperature the intrinsic thermal resistance of a crystal is set mainly by phonon-phonon Umklapp scattering: three-phonon collisions in which the crystal momentum changes by a reciprocal-lattice vector and heat-carrying phonon momentum is destroyed. In layered carbon the heat is carried disproportionately by low-energy acoustic phonons, including the flexural out-of-plane acoustic branch (usually called ZA), whose weak Umklapp scattering underlies the anomalously high conductivity of these materials.

Graphene, a single atomic plane of carbon, is the ultimate two-dimensional limit of graphite. Measured by a non-contact Raman optothermal technique - the sample is heated by a laser while its local temperature is read from the shift of the Raman G peak, and $K$ is extracted from the dissipated power, the temperature rise, and the flake geometry through a finite-element solution of the heat-diffusion equation - suspended single-layer graphene shows $K\approx 3000$-$5000$ W/mK near room temperature, well above the $\approx 2000$ W/mK of the basal planes of bulk graphite. This inversion of the usual thin-film trend had been predicted theoretically for strictly two-dimensional anharmonic crystals, whose intrinsic conductivity is expected to grow (logarithmically) with system size because Umklapp scattering of low-energy phonons is quenched; the phonon mean free path in graphene near room temperature is of order 800 nm, so flakes of several micrometers are already in the diffusive regime.

Few-layer graphene (FLG) - a stack of $n$ atomic planes, each about 0.35 nm thick, usually in Bernal AB stacking (successive planes occupying alternating A and B sites of the honeycomb lattice, as in common graphite) - is the natural system in which to watch heat conduction cross from two-dimensional to three-dimensional behavior. Ghosh et al. (2010) measured suspended, mechanically exfoliated FLG with $n=2$, 3, 4 and about 8, using the Raman optothermal method with per-sample finite-element extraction (identical flake geometry across $n$ cannot be obtained by exfoliation). They found $K$ falling from roughly 2800-3000 W/mK at $n=2$ to 1300-1500 W/mK at $n=4$, with the graphite value recovered at $n\approx 8$. They explained the evolution as an intrinsic effect: adding layers folds the phonon spectrum into $6n$ branches and couples the low-energy phonons across the planes, which enlarges the phase space available for three-phonon Umklapp scattering faster than it adds heat-carrying channels; first-principles theoretical values for single- and bilayer graphene based on this mechanism agreed with the experiment.

What that study itself left open frames the present problem. First, the precise layer number or structural condition marking the intrinsic crossover was not established. Second, the measured $K(n)$ carries extrinsic contributions - polymer residue from fabrication, thickness nonuniformity, edge roughness, sample-to-sample geometry, and heat-sink coupling - whose quantitative isolation was beyond the data (the quoted uncertainties from laser-spot size and Raman thermometry alone were 8-13%). Third, the theory included only three-phonon Umklapp and edge scattering, leaving higher-order anharmonicity, the temperature dependence of the crossover regime, and controlled disorder or strain untreated. Later work sharpened all three gaps rather than closing them: an independent thermal-bridge measurement of suspended FLG found room-temperature values of 389, 344, 302 and 596 W/mK for $n=2$, 3, 4 and 8 with no clear thickness dependence for the thinner samples; molecular-dynamics simulations found the interlayer registry effect on $K$ essentially unimportant while experiments on twisted and turbostratic multilayers found strong stacking dependence; and first-principles work showed that four-phonon scattering materially reduces even single-layer graphene's intrinsic conductivity. Understanding this crossover matters both for the physics of heat conduction in low-dimensional and van der Waals layered crystals and for the use of graphene films as lateral heat spreaders in electronics.

Problem Statement

Determine the intrinsic dimensional crossover of in-plane heat conduction in few-layer graphene. Concretely: (1) establish the room-temperature in-plane thermal conductivity $K(n)$ of suspended few-layer graphene as an intrinsic materials property - its values across $n$ quoted under a consistent flake-width normalization, its saturation toward the bulk graphite basal-plane limit, and the layer number at which the crossover from two-dimensional-graphene-like to three-dimensional-graphite-like transport completes; (2) identify the structural condition (layer number, stacking order, and interlayer registry) and the phonon-scattering mechanisms (cross-plane coupling of low-energy phonons, three-phonon versus four-phonon Umklapp processes, edge and boundary scattering) that set the crossover; and (3) quantify and separate the extrinsic contributions - polymer residue, thickness nonuniformity, edge roughness, measurement-extraction artefacts tied to flake width and lateral size (the intrinsic size dependence itself being fixed by the component-(1) normalization), heating-spot size, and contact or heat-sink coupling - so that the established $K(n)$ reflects the intrinsic materials property. These three components jointly serve one objective: a quantitative, experimentally validated account of how and why in-plane heat conduction crosses from two-dimensional to three-dimensional behavior as few-layer graphene thickens. The mechanisms and extrinsic channels enumerated in components (2) and (3) define the required scope: each named item must be identified or separated, or its negligible, inseparable, or already-accounted-for status demonstrated. Determine, through a quantitative microscopic theory, the single intrinsic room-temperature $K(n)$ of suspended few-layer graphene - quoted under the component-(1) flake-width normalization - that accounts for the measured $K(n)$ of both reported datasets (Raman-optothermal and modified T-bridge) across the reported layer numbers ($n=2$, 3, 4, and the approach to the bulk limit), determining which features of the measured dependence - including any decline from $n=2$ to $n=4$ and any saturation toward the bulk graphite basal-plane value near $n\approx 8$ - are intrinsic rather than extrinsic, and accounting for the technique-dependent spread between those datasets rather than absorbing it into the intrinsic curve.

References

1. Suchismita Ghosh, Wenzhong Bao, Denis L. Nika, Samia Subrina, Evghenii P. Pokatilov, Chun Ning Lau, Alexander A. Balandin, 'Dimensional crossover of thermal transport in few-layer graphene', Nature Materials 9, 555-558 (2010). DOI: 10.1038/nmat2753, https://doi.org/10.1038/nmat2753 (author manuscript at arXiv:1003.5247, https://arxiv.org/abs/1003.5247)
2. Alexander A. Balandin, Suchismita Ghosh, Wenzhong Bao, Irene Calizo, Desalegne Teweldebrhan, Feng Miao, Chun Ning Lau, 'Superior Thermal Conductivity of Single-Layer Graphene', Nano Letters 8, 902-907 (2008). DOI: 10.1021/nl0731872, https://doi.org/10.1021/nl0731872
3. Dhruv Singh, Jayathi Y. Murthy, Timothy S. Fisher, 'Mechanism of thermal conductivity reduction in few-layer graphene', Journal of Applied Physics 110, 044317 (2011). DOI: 10.1063/1.3622300, https://doi.org/10.1063/1.3622300 (preprint at arXiv:1104.4964, https://arxiv.org/abs/1104.4964)
4. L. Lindsay, D. A. Broido, Natalio Mingo, 'Flexural phonons and thermal transport in multilayer graphene and graphite', Physical Review B 83, 235428 (2011). DOI: 10.1103/PhysRevB.83.235428, https://doi.org/10.1103/PhysRevB.83.235428
5. Wanyoung Jang, Zhen Chen, Wenzhong Bao, Chun Ning Lau, Chris Dames, 'Thickness-Dependent Thermal Conductivity of Encased Graphene and Ultrathin Graphite', Nano Letters 10, 3909-3913 (2010). DOI: 10.1021/nl101613u, https://doi.org/10.1021/nl101613u
6. Michael Thompson Pettes, Insun Jo, Zhen Yao, Li Shi, 'Influence of Polymeric Residue on the Thermal Conductivity of Suspended Bilayer Graphene', Nano Letters 11, 1195-1200 (2011). DOI: 10.1021/nl104156y, https://doi.org/10.1021/nl104156y
7. W. Jang, W. Bao, L. Jing, C. N. Lau, C. Dames, 'Thermal conductivity of suspended few-layer graphene by a modified T-bridge method', Applied Physics Letters 103, 133102 (2013). DOI: 10.1063/1.4821941, https://doi.org/10.1063/1.4821941
8. Jin-Wu Jiang, 'Registry effect on the thermal conductivity of few-layer graphene', Journal of Applied Physics 116, 164313 (2014). DOI: 10.1063/1.4900526, https://doi.org/10.1063/1.4900526 (preprint at arXiv:1405.3031, https://arxiv.org/abs/1405.3031)
9. Xiangfan Xu, Luiz F. C. Pereira, Yu Wang, Jing Wu, Kaiwen Zhang, Xiangming Zhao, Sukang Bae, Cong Tinh Bui, Rongguo Xie, John T. L. Thong, Byung Hee Hong, Kian Ping Loh, Davide Donadio, Baowen Li, Barbaros Ozyilmaz, 'Length-dependent thermal conductivity in suspended single-layer graphene', Nature Communications 5, 4689 (2014). DOI: 10.1038/ncomms4689, https://doi.org/10.1038/ncomms4689
10. Hongyang Li, Hao Ying, Xiangping Chen, D. Nika, Alexandr I. Cocemasov, Weiwei Cai, A. Balandin, Shanshan Chen, 'Thermal conductivity of twisted bilayer graphene', Nanoscale 6, 13402-13408 (2014). DOI: 10.1039/c4nr04455j, https://doi.org/10.1039/c4nr04455j
11. Denis L. Nika, Alexander A. Balandin, 'Phonons and thermal transport in graphene and graphene-based materials', Reports on Progress in Physics 80, 036502 (2017). DOI: 10.1088/1361-6633/80/3/036502, https://doi.org/10.1088/1361-6633/80/3/036502 (preprint at arXiv:1606.00488, https://arxiv.org/abs/1606.00488)
12. Tianli Feng, Xiulin Ruan, 'Four-phonon scattering reduces intrinsic thermal conductivity of graphene and the contributions from flexural phonons', Physical Review B 97, 045202 (2018). DOI: 10.1103/PhysRevB.97.045202, https://doi.org/10.1103/PhysRevB.97.045202
13. Shuo Han, Xianhua Nie, Shangzhi Gu, Wenyu Liu, Luchen Chen, Hao Ying, Le Wang, Zhihai Cheng, Li Zhao, Shanshan Chen, 'Twist-angle-dependent thermal conduction in single-crystalline bilayer graphene', Applied Physics Letters 118, 193104 (2021). DOI: 10.1063/5.0045386, https://doi.org/10.1063/5.0045386
14. A. Mohapatra, M. S. Ramachandra Rao, Manu Jaiswal, 'Thermal transport in turbostratic multilayer graphene', Carbon 201, 120-128 (2023). DOI: 10.1016/j.carbon.2022.08.089, https://doi.org/10.1016/j.carbon.2022.08.089 (preprint at arXiv:2206.04637, https://arxiv.org/abs/2206.04637)
15. Zherui Han, Xiulin Ruan, 'Thermal conductivity of monolayer graphene: Convergent and lower than diamond', Physical Review B 108, L121412 (2023). DOI: 10.1103/PhysRevB.108.L121412, https://doi.org/10.1103/PhysRevB.108.L121412 (preprint at arXiv:2302.12216, https://arxiv.org/abs/2302.12216, under the earlier title 'Is Thermal Conductivity of Graphene Divergent and Higher Than Diamond?')
16. Jiawei Li, Jie Zheng, Linxin Zhai, Peng Liu, Fangzhu Qing, Xuesong Li, Di Chen, Zhiping Xu, Haidong Wang, Hongwei Zhu, 'Layer-Dependent Thermal Conductivity of Graphene toward Thermal Management Applications', ACS Applied Materials & Interfaces 18, 11764-11773 (2026). DOI: 10.1021/acsami.5c24572, https://doi.org/10.1021/acsami.5c24572